The present invention relates generally to packaged electronic sensors, and more particularly to packaged electromagnetic type pickups, also known as magnetic pickups, magnetic sensors, speed sensors, or more accurately variable reluctance sensors. Such sensors are used to measure the rotational speed or the position with respect to a fixed reference of a ferrous exciter rotor or reluctor wheel.
Variable reluctance speed and position sensors are well-known and widely used in automotive applications. A general theory of variable reluctance sensor operation is explained in, for example, U.S. Pat. No. 3,614,615, issued to Cass, and U.S. Pat. No. 4,045,738, issued to Buzzell.
The early automotive applications of variable reluctance sensors were for inputting of speed information for electronic speedometers and tachometers on heavy and medium duty trucks and off-road equipment (agricultural, construction, and recreational). The speed sensors were also used to supply wheel speed information for antiskid braking systems for heavy trucks. In general the sensors comprised a plastic bobbin which housed the magnetic elements (pole piece and magnet) and magnet wire. The bobbin assembly was encapsulated into a metal or plastic housing via injection molding of thermoplastic resins or via transfer molding of thermosetting resins, or the encapsulating material itself served to house the bobbin assembly. The encapsulating process frequently caused damage to the magnet wire which resulted in an "open circuit" failure condition early in field service as temperature cycling caused the damaged magnet wire to break (open circuit).
More recently, applications for variable reluctance sensors have increased. With the advent of more aerodynamic vehicles and the use of four and six cylinder engines at higher rpm's to attain power levels of eight cylinder engines as well as to achieve improved fuel economy, the under hood temperatures are higher. The mounting locations for these sensors in the new applications place them in transmission fluid, engine oil, hydraulic fluid, and salt water which are corrosive materials or carriers of corrosive agents used in the manufacture (machining) of metal parts. Together with longer warranty objectives, these harsher environmental conditions have required more difficult and more accelerated life testing for sensor design validation.
As automotive O.E.M.'s continue to look for ways to meet government mandates for emission levels and fuel consumption and to provide improved vehicle performance and safety to consumers, usage of variable reluctance sensors is mushrooming. Some applications include but are not limited to: direct sensing crankshaft position for spark timing in distributorless ignition systems; direct sensing camshaft position for timing of sequential fuel injection systems; speed input of information for antilock braking systems, electronically controlled transmissions, automatic all wheel drive systems, traction control, as well as the previously mentioned input speed information for instrumentation (vehicle speed and engine speed); and direct sensing of camshaft position for variable camshaft timing systems designed to improve engine horsepower while simultaneously reducing emissions and fuel consumption.
A need for improved sensor reliability and design life is necessitated by the ever increasing warranty periods with targets for warranties out to 100,000 to 200,000 miles as well as by the critical nature of functions performed on vehicles by the sensors as described above. Failure of the variable reluctance sensor will result in vehicle non-operation, safety risk, or poor engine performance in emission control or fuel consumption.
As the warranty periods and use of variable reluctance sensors for critical functions increase, the latent failure modes as well as early failure modes of variable reluctance sensors must be addressed. All the while, the environmental conditions imposed upon these sensors are becoming more demanding.
Further, the output (threshold voltages) requirements are increasingly more difficult to meet because the package size available for the sensors is shrinking with smaller and lighter vehicles. Also, with emission requirements becoming tighter and even required during engine start-up, ignition timing must be achieved during engine start-up which means the same sensor output must be attained at lower exciter gear speeds. This translates to less propensity to induce voltage since the lower the speed of the exciter gear the less the propensity of a given system to induce voltage in a given sensor.
Together with the more critical nature of the applications and the harsher environments, much work has been done to reduce the encapsulation damage to the magnet wire, such as lower molding pressures and/or a provision of protective layering or shielding between the coil and encapsulant. Progress has been made in reducing failure rates; however, the open circuit condition still exists as a result of differing thermal expansion rates of the encapsulant and magnet wire or as a result of the transmission of encapsulant pressure damage through or around the protective layer.
Sealing from liquid intrusion must be provided at all points or surfaces on a variable reluctance sensor including the back and rear end (cable or connector end) of a sensor in applications where complete submersion is possible or where corrosive liquid splash or spill is possible. Many approaches have been taken to seal off the intrusion of corrosive liquids into the sensor. None have been successful in solving the open circuit condition resulting from encapsulation/thermal cycling damage and the sealing off of corrosive liquid intrusion, without sacrificing the output capability of the sensor.
In U.S. Pat. No. 4,680,543, issued to Kohen, a sleeve-like molded plastic cover surrounds the spool (coil), and the magnetic element abuts the inner surface of the cover wall. Further, an encapsulation occupies the spaces within the cover. The thickness of the cover wall increases the working air gap between the exciter rotor and the sensor magnetic element thereby reducing output capability. Also, the sensor coil is still subject to encapsulation/thermal cycle damage to the magnet wire coil. Lastly, although the cover seals off liquid intrusion at the sensing side of the sensor, the design does not: provide for an adequate seal from liquid intrusion via the cable.
In U.S. Pat. No. 4,847,557, issued to Saito, et al., the sensor element is covered by a first thermosetting resin inside a case, wherein a second thermoplastic resin hermetically seals the opening in the case. One of the problems with this design is that it does not enable the use of a high coercive force rare earth magnet and thereby does not maximize the output capacity of the sensor. Further, encapsulants are utilized to form seals. In addition to causing magnet wire damage (sensor failure) during thermal cycling, encapsulants have proven to be an ineffective sealing medium during thermal shock or thermal cycles/submersion testing. In other words, this sealing approach has not been sufficient to eliminate intrusion of corrosive liquid into the sensor because the differing thermal expansion rates of the different materials cause any encapsulated material adhesion (sealing) to separate during thermal cycling, thereby creating a leak path.
Although Saito, et al. attempts to employ a seal ring between the bracket and the housing to overcome the separation phenomenon described above, the seal ring only provides a seal for one of three leak paths. The remaining two leak paths are still subject to the separation problem in thermal cycling: 1) between the output wire protective member and the second resin member, and 2) between the magnetic pole and the case, further between the bobbin and the case, and further between the bobbin and first resin member. Penetration of corrosive liquids via the leak paths will result in a shorted coil, resulting in sensor failure.
Consequently, a need exists for a variable reluctance sensor capable of high reliability, not subject to encapsulation damage and impervious to liquid intrusion while maximizing output capability per sensor package volume.